September 5, 2006 - Observatory Test Planning meeting

J Bogart, R Claus (chair), K Fouts, E Grove, P Hart, N Johnson, B Lee, J Panetta, J Thayer, S Tuvi

FSW (J Thayer)

Build B0-7-0 is targeted for 9/15
- Major change affecting LICOS et al. is mnemonic changes
Planning 2 days of regression testing
Planning a week of test bed time for testing with LICOS scripts
- Mostly involves intSEApp
- Also power On/Off scripts are affected
Most changes are already in the prod branch in AFS
Upload to LAT is planned for 9/25, followed by regression testing on the LAT through 9/28
- Needs to be coordinated with LICOS, et al. releases
- Goal: Make beta release of LICOS, et al. by 9/15

New Scripts

LAT-12x (Science Modes)
- Eric indicated that this must be completed no later than 9/25 (P Hart)
- Window for execution on the LAT is 9/25-28
LAT-20x (Science Performance Diagnostics)
- Additional configurations in E2E csv file
- Completed
LAT-23x (GRB Handling)
- Update needed to use new functionality provided by FSW B0-7-0
- Philip estimates it's a couple of days of work
- Eric points out this script will be eclipsed by a test to interact with the spacecraft and the GBM, so it is lower priority to get this done
- Neil says that perhaps the FQT is adequtate to demonstrate the FSW functionality
- The most important task here is to develop a plan for the Observatory-level test, then provide to / coordinate with Chip Meegan (GBM at MSFC) to agree on details
LAT-90x (File system dump)
- Needs to be run prior to LAT EGSE shipment to SASS (9/12)
- R Baun to supply details

Tools for generating configurations (E Grove)

Joanne is working on tools related to this
- Won't be ready on the one month time scale
Byron has been working such tools as well
- Tool is meant to be used to populate MOOT
- It generates configurations from scratch
- He could return to working on this after the bent pipe work is done
- Limited amount of time: he'll be leaving GLAST at the end of this month
- So far only CAL, ACD and Trigger are supported
Need to set up a brainstorming session to define what we want
- Interface - How to interact with the tool(s)
- Nomenclature for the intent of a configuration
Post meeting discussion with Jim and Selim:
- Jim has started work on a tool chain that reads a LATc base config file and outputs a LATc config file, recalculating broadcasts
- Could add folding in of 'delta' config files to change things like the bad channel list
- This tool would address modifying existing LATc configurations
- It avoids the time consumming LATTE based approach Lester set up
- Jim agreed to write a note in Confluence describing the tools and how a user interacts with them

New Analysis Engine (P Hart)

Key change is the new version runs on datagram (LSF) files instead of LDF files
System is robust against dropped datagrams
System automatically processes new data files and retires them if the analysis script ran successfully
Updated GUI showing node running the analysis and the elapsed time
System is working with the exception of for the TkrNoiseAndGain and TkrTotGain scripts
- TkrNoiseAndGain takes too long (24 hrs) and runs out of memory
- TkrTotGain has a JIRA open against it LS-108 (also Jim says LCS-117 is still an issue)
Philip would like help to verify script outputs

Data Ingest

MOOT (J Panetta)

Jim has code base mostly ready for LICOS
Need to import config files into MOOT
Need to figure out how to label things and codify intent
- Eric says that a small number of runs should cover most of the space
- Eric agreed to make a list to help get this process started

See "Tools for generating configurations" above
Jim to work on integrating latest MOOT release into LICOS
- Goal is to include this in the next LICOS release
- Will update a couple of scripts to use MOOT

Ancillary Tracking DB tool (B Lee)

Tool is ready
Need a good, solid Ancillary area snapshot to populate the database
- Only CAL writes a pedestal file during the CPT, else Ancillary isn't modified by CPTs
Pick up the snapshot during shipment of the LAT (between pre- and post-ship CPTs)
Eric says we'll want to recalibrate after TVAC
Need to carefully coordinate taking the snapshot in order to not get confused by running different configurations for pre- and post-ship tests

There is a lot of work to do in 3 weeks
- Not listed above: Bent Pipe ramifications, lots of outstanding JIRAs
Push for getting new Analysis Engine and Data Ingest deployed by 10/1
Get LAT-12x done by 9/25
Get LAT-90x done by 9/9